On Friday, January 06, 2006 4:57 PM, Radu Preotiuc-Pietro <mailto:[EMAIL PROTECTED]> wrote:
> I would say if that data is well-formed XML then your > original approach > would work, with the possible addition of "mixed=true" if > your client's > content can have text at top-level. > Thank you very much for your reply. The data is an XML fragment that consists of one or more well-formed elements. It originally arrives as the contents of a blackBox element in the input. The input process just grabs the elements as a String and stores them in a database. > Then the "XML generator" that you are talking about should get a clue > probably based on the name of the type/element that it writes out > (possibly the contents of that type but that may have > side-effects that > you don't expect in case you have other types with similar > content) that > it is the "black box" content and then write it out. It seems > to me that > the generator is likely the problem in this case. > That is the tack I am taking so far (taking special action for elements named "blackBox"), but I still have a problem. I am using a StAX XMLOutputStream to write the data, through its writeCharacters method. It is "helping" me by wrapping the whole thing in a CDATA section to save me from the illegal characters in the text. This, of course, has nothing to do with XMLBeans. -- Jim McMaster mailto:[EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
